IXTEODUCTION. 17 



coTiiTnunicated to the air lias been supposed to con- 

 stitute a wind blowing in the direction of the spot 

 where the heat was generated. It must he confessed, 

 however, that the cause thus adduced does not seem 

 adequate to produce the effects attributed to it,* 

 though probably some of the currents of the air are 

 owing to variations of its temperature. And as these 

 variations are perpetually occurring, dependent on 

 causes which are difficult to detect, and as the aerial 

 currents resulting from them act and react on each 

 other, variously modifying their direction, force, and 

 duration, the ordinary winds are irregular and incon- 

 stant even to a proverb. Some observations, however, 

 recently made, have revealed some particulars of a 

 highly-interesting character, concerning the winds of 

 the temperate zones : one of which is, that they blow 

 in a circular direction ; that is, the course which a 

 storm has taken, if marked out on a map or globe, 

 would describe a circle, often of many degrees in 

 diameter. The direction of the gale in the circle is 

 not arbitrary, but seems to be invariably from north 

 to west, south, and east, in the northern hemisphere, 

 and in the opposite course in the southern. These 

 winds appear to be intimately connected M'ith mag- 

 netism : it is a curious fact, that, in the midst of the 

 southern Atlantic, where magnetic influence is at the 

 lowest degree of intensity, storms are unknown, while 

 the meridians of the magnetic poles, that of the Ame- 

 rican cutting the West Indies, and that of the Siberian 

 the China Sea, are peculiarly liable to tempests ; the 

 hunicanes of the former, and the typhoons of the 

 latter, being well known.* It is pretty certain, also, 



* See Eeid on Storma. 

 C 



